Introduction

Trypanosomosis is a widely distributed parasite-associated disease that affects humans and animals. Trypanosoma vivax cause significant economic losses in livestock, being responsible for major losses in sub-Saharan Africa and Central and South America (^{DÁVILA & SILVA, 2000}). Infections by T. vivax in livestock have been occurring with increasing frequency in tropical and subtropical regions (^{DÁVILA & SILVA, 2000}; ^{GIORDANI et al., 2016}). In cattle, T. vivax diagnosis is difficult, since the parasite triggers non-specific symptoms such as fever, anorexia, weight loss, reduced milk production, abortions, and neurological signs (^{BATISTA et al., 2007}; ^{CADIOLI et al., 2012}). Moreover, there are fluctuations in parasitaemia and apparently aparasitemic intervals (^{DESQUESNES, 2004}; ^{CADIOLI et al., 2015}; ^{FIDELIS et al., 2016}).

Direct parasitological methods present poor sensitivity during low parasitaemia or aparasitemic periods, which are common in chronic infection (^{DESQUESNES, 2004}; ^{CADIOLI et al., 2015}). Serological tests, including indirect immunofluorescent antibody test (IFAT), enzyme-linked immunosorbent assay (ELISA), and immuno-chromatographic rapid diagnostic test (RDT), are the tools of choice for herd screening but do not indicate if the infection is active or if the animal has responded to treatment (^{CADIOLI et al., 2012}, ²⁰¹⁵; ^{BOULANGÉ et al., 2017}). On the other hand, molecular methods such as conventional polymerase chain reaction (PCR) and real-time PCR (qPCR) have the potential to be excellent diagnostic tools; however, these techniques are subject to false negative results when the parasitaemia is below the detection level (^{CADIOLI et al., 2015}; ^{KATO et al., 2016}).

For trypanosomiasis control, integrated approaches such as vector control, improved general health and immunocompetence of affected herds, and the use of more sensitive and rapid diagnostic tools are necessary (^{GIORDANI et al., 2016}). Thus, the present study aimed to evaluate the use of several techniques for T. vivax diagnosis during the course of the disease in order to assess the best way of use them.

Materials and Methods

Results

The results are presented in Table 1 and Figure 1 and show the presence of trypomastigotes from the second post-infection collection (7 DAI) as verified by the three parasitological methods. Among the methods used for direct parasite detection, the hematocrit centrifugation technique had the highest detection capability, detecting 24 of 54 known positive samples (44.4%). However, the thick-drop and stained blood smear techniques showed similar detection capabilities, detecting 17 of 54 known positive samples (31.5%) (Table 1).

Figure 1 Parasitemic curves quantified by qPCR and relative intensity of PCR of three individual cows (E1, E2 and E3) experimentally infected with T. vivax. 

The IFAT and ELISA techniques performed comparably with the IFAT being slightly more sensitive, detecting seropositivity in 51 of the 54 known positive samples (94.4%). From 7 DAI, all animals showed reactivity at the 1/80 serum dilution and after 21 DAI achieved seropositivity at the 1/1280 dilution, maintaining this level (or higher) until the end of the experimental period. The exception was the E1 animal, whose titre dropped to 160 at 119 DAI. The ELISA test detected seropositivity in 49 of the 54 known positive samples (90.7%). At 7 DAI only one animal (E1) was seropositive, but from 14 DAI, all samples were seropositive and remained so throughout the experiment (Table 1).

Regarding the molecular evaluation, all samples were positive for the host endogenous gene TLR-8, indicating that extractions were efficient, and showing that the results obtained by both PCR and qPCR are reliable. The mean and intervals for efficiency, R², slope, and y-intercept of qPCR reactions were 93.6%, (90.1–97.1), 0.988 (0.970–0.999), –3.487 (–3.586–(–)3.392) and 39.758 (38.583–41.764), respectively. All duplicates presented a maximum variation of 0.5 Cq. Both techniques detected the T. vivax DNA in 33 of the 54 known positive samples (61.1%) with the first detection on the first day after infection (1 DAI) (Table 1 and Figure 1).

The highest values of parasitaemia, detected by the thick-drop technique, occurred on 14 DAI with 6.82 × 10⁶, 1.62 × 10⁷ and 3.99 × 10⁶ parasites/mL of blood for E1, E2 and E3 cows, respectively. The quantification by qPCR also revealed that the highest parasite values occurred on 14 DAI, showing 5.03 × 10⁷, 2.70 × 10⁸ and 8.53 × 10⁷ copies of the target region/mL of blood, for E1, E2 and E3 animals, respectively (Table 1 and Figure 1). Fluctuations in parasitaemia were observed during the experimental period both by qPCR and thick-drop techniques, highlighting a patent phase, in which circulating parasites were seen throughout the entire period, and subpatent phase, in which a low parasitaemia was detected followed by aparasitemic periods. The parasitemic curves quantified by the qPCR technique are shown in Figure 1, while the parasitemic curves quantified by thick-drop technique were reported in ^{Fidelis et al. (2016)}.

Kappa agreement results are presented in Table 2. Both correlation analyses were shown to be positive according to Spearman r = 0.8757 (P <0.0001) for the two quantitative techniques, thick-drop technique and qPCR, and Spearman r = 0.9151 (P <0.0001) for the qPCR quantification and PCR relative intensity.

Discussion

South American isolates of T. vivax do present some genetic differences despite being phylogenetically related to African isolates (^{GARCIA et al., 2014}). The present study was the first to use the qPCR protocol, developed for African isolates (^{SILBERMAYR et al., 2013}), for one South American isolate, detecting the T. vivax DNA, “Lins” isolate. Clearly it will be possible to develop a rapid and accurate measurement of parasite numbers in whole blood through target DNA quantification.

The short pre-patent period verified in the present study, in which molecular methods were used, may have occurred by the highly inoculum concentration employed. This highly concentration was similar to those ones employed in the studies performed by ^{Igbokwe et al. (1996)} and ^{Schenk et al. (2001)}. It was also observed that the PCR assay possessed an identical capacity of detection for T. vivax DNA as the qPCR assay (61.1%). This similarity in detection capacity could be explained by the fact that the qPCR assay described by ^{Silbermayr et al. (2013)} was directed to ITS1 region that has a high CG content, which can interfere with the diagnostic accuracy of qPCR (^{FIKRU et al., 2016}). Fluctuations in parasitaemia and aparasitemic intervals make the diagnosis of animal trypanosomiasis challenging since it is difficult to directly detect the parasite, especially in the subpatent phase of infection (^{CADIOLI et al., 2015}); thus, the use of more sensitive diagnostic tools such as PCR and qPCR is necessary. The detection rate of the molecular methods was higher than the parasitological methods. Both molecular techniques were able to identify 61.1% of the known positive samples, whereas the hematocrit centrifugation technique, the best parasitological method, detected 44.4%. The molecular detection rates are similar to that described by ^{Cadioli et al. (2015)}, who also used the T. vivax CatL target region and worked with experimentally infected cattle. Previously ^{Tran et al. (2014)} using the ITS1 as target region when working with experimentally infected cattle, obtained a detection rate of 84.9%.

Through qPCR analysis it was possible to verify that the patent phase of the illness, where parasites can be detected without interruption, may extend up to 42 DAI. This is well beyond the 14 DAI and 30 DAI previously reported by ^{Rodrigues et al. (2013)} and ^{Fidelis et al. (2016)}, respectively, which used direct parasitological techniques to quantify the parasitaemia. Thus, the use of more sensitive molecular tool may help in a better understanding of the parasite dynamics during the infection. The two molecular techniques presented similar detection rates, obtaining a Kappa index of 0.87, which presented a strong agreement (^{LANDIS & KOCH, 1977}). Although molecular techniques are effective in detecting infected animals in the patent phase of the disease, they have failed, on some occasions, to detect known positive samples during the subpatent phase. Serological tests can be useful tools for this phase of the illness, especially for cases that occurred in areas considered disease free.

High antibody levels have been described as an efficient mechanism to control parasitaemia (^{MATTIOLI & WILSON, 1996}), which may explain the findings of the present study where all animals were seropositive from 14 DAI, for both serological techniques. The higher sensitivity of serological tests when compared to parasitological (^{MATTIOLI et al., 2001}) and molecular tests (^{CADIOLI et al., 2015}) suggests their use in epidemiological studies, in which the overall exposure to trypanosome infection is being investigated, and/or in cases in which no treatment was performed or non-specific anti-Trypanosome drugs were used. Care should be taken when using a single ELISA for T. vivax diagnosis, as antibodies are likely to cross-react with other parasites of the same genus (^{DESQUESNES, 2004}). Another complicating factor is that serological tests do not indicate if the infection is active or if the animal has responded to treatment (^{CADIOLI et al., 2012}, ²⁰¹⁵; ^{BOULANGÉ et al. 2017}). According to ^{Batista et al. (2007)}, anti-T. vivax antibodies in cattle that did not present clinical signs suggest the occurrence of subclinical disease, a form of the disease in which the detection of circulating parasites is difficult.

Although T. vivax visualization in blood smears is still the main method for the diagnosis of the disease and direct parasitological techniques are widely practiced (^{DESQUESNES, 2004}), these methods presented the lowest sensitivity among the tests evaluated. Serological techniques are efficient in detecting seropositive animals, especially in the subpatent phase of infection, but in spite of being highly sensitive, they only demonstrate exposure to the infectious agent (^{DESQUESNES, 2004}; ^{OSÓRIO et al., 2008}). The molecular techniques are highly sensitive, especially in the patent phase of the disease, but also demonstrated good sensitivity during the subpatent phase. They could be used in conjunction with serological techniques in the search of animals that remain infected after treatment (unpublished data). With the emergence of more sensitive and accurate molecular tools we anticipate that in the future the detection of positive animals will be improved, even in cases of low parasitaemias. However, for the present time, the combination of several diagnostic techniques can avoid the generation of false negative results, enable a better disease control and, consequently, the reduction of economic damage generated by T. vivax infection of productive cattle.